By Ross Fisher
Predictive modeling is transforming how claim examiners identify and manage high-risk workers' compensation claims, and yielding results that are helping to reduce return-to-work time frames and claim costs. Vice President and Claim Data Scientist Paul Drennan is a driving force behind The Hartford's innovative approach to predictive modeling, which informs the company's comprehensive solutions for workers' compensation claim intervention and management. Paul and I recently sat down to discuss how predictive modeling supports the claim handler, and why it is such a critical value proposition for large, loss-sensitive companies.
RF: As we've discussed in prior articles, insurance companies are increasingly using predictive modeling to forecast future outcomes based on past experience. How is predictive modeling in use at The Hartford?
PD: A primary application of predictive modeling at The Hartford is in the area of workers’ compensation claims. As far back as 2008, we deployed a model to identify potentially high-risk workers' compensation claims before they become volatile. The model that we launched has been rebuilt and improved upon twice, and is patented.
RF: How does predictive modeling support the workers' compensation claim handler?
PD: Predictive modeling provides claim handlers with objective data that aids their decision-making and ultimately helps to improve claim outcomes. Our model watches the details of a claim on the handler’s behalf, providing insight that enables the handler to intervene proactively before conditions or costs escalate.
Sometimes a claim is so unique that signs of volatility stand out right away. The claim handler doesn't necessarily need a predictive model to detect these obvious outliers. Where the model delivers its greatest benefit is for claims that are more nuanced and less commonplace, and therefore the handler may be less tuned into their potential for volatility.
RF: Could predictive modeling ever replace the role of the claim handler?
PD: Both data and people are critical to success – and at The Hartford, the claim handlers are primary. Our objective is to wed the qualities of great claim handlers with the tools that allow them to do their jobs better and more efficiently. The predictive model is just one of the tools within the handler's toolbox.
RF: What specifically human qualities are essential to the claim handling process?
PD: All claims begin with the connection that's established among the claim handler, claimant and other specialized resources that are brought in to support the employee's return to health and productive work. Rapport among all parties is essential, and it builds over the course of a healthy claim. That's an exclusively human thing; the data and algorithms that feed our model have nothing to do with rapport.
Effective claim handling also depends upon the handler's ability to collect all necessary data, navigate the organization to bring in supporting resources and make effective claim decisions. This expertise doesn't go away.
RF: How does predictive modeling improve the claim handler's performance?
PD: An effective model puts every handler's performance on a par with the best. It studies what top claim handlers do, looks for what they look for, systematically finds what they find and alerts everyone in the claim process who needs to know. The model is also consistent. Even the best handlers aren't at the top of their game all the time, but the model never has a bad day.
We've also trained our model to detect symptoms of lack of rapport, whether it was lost at some point in the claim process or never established in the first place. A handler who is alerted to this potentially negative sign is better positioned to restore communication among all parties, address any issues and get the claim back on track.
RF: Can you provide an example of how predictive modeling can augment a claim handler's expertise?
PD: Let's take the case of dispensing opioids. We rank prescription providers on how quickly they start to prescribe opiates and in what proportion, which allows the model to issue an alert when atypical behaviors are detected. A claim handler who has never worked with a particular physician or pharmacy might not suspect a risky dispensation pattern but the model would quickly spot any variances and deliver the insight the handler needs to take effective action.
RF: How does predictive modeling fit into the day-to-day life of a claim handler?
PD: Our model is integrated into the systems that the claim handlers use to do their jobs every day. The claim handler is not aware of the predictive model per se, but will see the fruits of its work in the form of a list of alerts that have been issued to different claim teams, a message about a specific claim, or a notification that a claim has been referred to fraud handlers or some other area. The handler may choose to use the information provided by the model or not.
RF: Sometimes change in process is disruptive but this sounds empowering. How are handlers responding?
PD: Qualitative feedback has been positive. Here's what a few of our handlers have had to say:
“The claim indicators are a helpful tool to trigger us as claim handlers to investigate further."
“Receiving a new claim comes with a flurry of work. It's nice to know that our claim system is looking for red flag indicators to automatically refer my claim to the subrogation or special investigations unit if needed."
RF: What kind of training do claim handlers receive regarding predictive modeling?
PD: The Hartford has a strong commitment to training. Our claim handlers go through an intense training period before they ever handle a claim, and we offer upskilling opportunities after initial training in the form of formal classroom or lab experiences. We also encourage less formal continuous learning through badging, a process that incents staff to build skills on their own for which they are awarded a badge as tangible recognition and further credentialing. Data is one of the areas in which handlers can earn different levels of badges. We encourage their knowledge of data science.
RF: How does the human-data relationship translate to improved outcomes for The Hartford's customers?
PD: Data helps the claim handler work more efficiently and achieve better claim outcomes. Ultimately, this means injured employees can get the care they need and return to the life they had before their injury. The employer benefits by getting their trained staff back to work sooner, and earlier closure rates for their workers compensation claims lead to a lower cost of risk.
About Paul Drennan
Paul Drennan is vice president and predictive analytics lead in the data science unit at The Hartford. With more than 20 years of experience establishing and leading data science teams across multiple industries, Drennan focuses on marrying world class analytics with deep process knowledge in large scale operations to create sustained profitable changes in business practices.
The information provided in these materials is intended to be general and advisory in nature. It shall not be considered legal advice. The Hartford does not warrant that the implementation of any view or recommendation contained herein will: (i) result in the elimination of any unsafe conditions at your business locations or with respect to your business operations; or (ii) will be an appropriate legal or business practice. The Hartford assumes no responsibility for the control or correction of hazards or legal compliance with respect to your business practices, and the views and recommendations contained herein shall not constitute our undertaking, on your behalf or for the benefit of others, to determine or warrant that your business premises, locations or operations are safe or healthful, or are in compliance with any law, rule or regulation. Readers seeking to resolve specific safety, legal or business issues or concerns related to the information provided in these materials should consult their safety consultant, attorney or business advisors.
The Hartford® is The Hartford Financial Services Group, Inc. and its subsidiaries, including Hartford Fire Insurance Company. Its headquarters is in Hartford, CT.